Influenza may jeopardize the health of patients with diabetes mellitus in several ways. In the first place influenza infection may inbalance a carefully established metabolic control, and in some cases trigger a process of metabolic deterioration which eventually may lead to ketoacidosis and even death (40-42). Secondly, diabetes itself might be the cause of an impaired immune response to influenza virusses. Patients are made more vulnerable to infection, especially if they are in poor metabolic control ( 46,4 7). In the third place pre-existing staphylococcal skin infections can enhance the incidence of the most dreaded complication of influenza infection: secondary staphylococcal pneumonia (32-36). · An increased carrier rate of S. aureus in combination with an impaired immune response to this microorganism can be held partly responsible for the increased morbidity and mortality in patients with diabetes mellitus. Patients who have overt skin lesions should receive anti-staphylococcal antibiotic therapy as soon as symptoms of influenza infection are observed. Though annual vaccination has been proven to reduce attack rates and alleviate illness (48) many patients with diabetes mellitus are still not vaccinated against influenza. Physicians who deny the need for annual vaccination argue that the excess mortality from influenza in patients with diabetes mellitus dates back from earlier days when patients were not very well controlled. From their point of view there is no need for mass vaccination of patients who are nowadays mostly well controlled. In our opinion there is no reason for such optimism. In both clinical studies and studies on excess mortality diabetes mellitus is a remarkably constant risk factor over a long time. Though it is difficult to calculate reliable figures on relative risks and rates of excess mortality there is sound evidence to assume that in epidemic periods mortality in patients with diabetes mellitus increases by 5-15% (9,11). If one considers the enormous effort that is made to attain satisfying metabolic control and to fight the secondary complications of diabetes a single injection once a year to protect against influenza is not overdone. Aside from annual vaccination of all patients with diabetes mellitus (both type 1 and type 2) we propose that in patients with additional risk factors the response to vaccination be monitored with standard sero-logical methods (haemagglutination inhibition or single radial haemolysis). Patients that remain unprotected after vaccination should receive amantadine 200 mg a day during an eventual epidemic, which may protect against type A but not type B infection (49).